foresight_smooth

foresight_smooth, gnuplot smooth filters of a data series.

The filters gnuplot computes from the data alone, without a fit: each run of placeable points (a block of the data file, broken by undefined points and by points not placeable on a log axis) is sorted by x, stably, and its points of equal x are merged into one:

• unique: the mean of their y;
• frequency: the sum of their y; fnormal divided by the sum of y over every run;
• cumulative: the sum of y up to that x within the run; cnormal divided by the sum of y over every run.

Runs stay separate, joined by an undefined (NaN) point that breaks the line, as gnuplot draws them.

Source: src/lib/foresight_smooth.F90

Dependencies

Contents

• smooth
• placeable
• sorted

Variables

Name	Type	Attributes	Description
SMOOTH_MODES	character(len=*)	parameter	Supported filters.

Subroutines

smooth

Filter the series (x, y) with the smooth mode (one of SMOOTH_MODES) into (xs, ys); xlog, ylog tell the log axes, on which non-positive values are not placeable.

subroutine smooth(mode, x, y, xlog, ylog, xs, ys)

Arguments

Name	Type	Intent	Attributes	Description
mode	character(len=*)	in		Filter.
x	real(kind=R8P)	in		Abscissae.
y	real(kind=R8P)	in		Ordinates.
xlog	logical	in		Log x axis.
ylog	logical	in		Log y axis.
xs	real(kind=R8P)	out	allocatable	Filtered abscissae.
ys	real(kind=R8P)	out	allocatable	Filtered ordinates.

Call graph

Functions

placeable

Whether v is placeable on an axis: finite, and positive on a log axis (as foresight_axis%accepts).

Attributes: elemental

Returns: logical

function placeable(v, log) result(ok)

Arguments

Name	Type	Intent	Attributes	Description
v	real(kind=R8P)	in		Value.
log	logical	in		Log axis.

Call graph

sorted

Permutation sorting keys ascending, stable (equal keys keep their order): bottom-up merge sort.

Attributes: pure

Returns: integer(kind=I4P)

function sorted(keys) result(order)

Arguments

Name	Type	Intent	Attributes	Description
keys	real(kind=R8P)	in		Keys, no NaN.

Call graph